Issues concerning the collection and uses of data about college student retention are considered. Some of the documented findings about student retention in U.S. colleges is summarized, with attention to both national data and data for colleges belonging to the United Negro College Fund. Difficulties in obtaining useful retention data are…

Noting that improvement in rater reliability means eliminating differences among raters, this paper discusses ways to assess writing evaluator reliability and methods for achieving higher levels of interrater reliability. After showing that reliability can be improved two ways--by increasing the number of raters or measurements made, and by…

Provides examples of the use of computer simulation to help beginners develop a grasp for difficult statistical concepts. The implementation of this teaching methodology is discussed and analysis of simulation output through use of standard statistical software packages is illustrated. (Author/MBR)

The mnemonic keyword study of Johnson, Adams, and Bruning (1985) on effectiveness of the keyword method for second-language vocabulary acquisition is critiqued. Of particular concern are literature currency vis-a-vis the authors' stated rationale for conducting the study, and methodological issues associated with the study's implementation.…

Examines difficulties associated with the use of conversational features in the analysis of mother-infant vocalization episodes. Observations of six mother-infant dyads revealed difficulties involving (1) pooling of data; (2) the unit of analysis; and (3) selection of appropriate "expected" values for dyadic parameters used in tests of…

Indiana University-Purdue University at Indianapolis (IUPUI) offers an annual 2-week intensive Summer Faculty Writing Forum. This Forum accepts up to 15 faculty each year from schools and disciplines across the campus. These faculty members, more used to the role of writing to demonstrate learning, investigate the capacity of writing to…

Since the latter part of the 19th century, a fervent debate has ensued about quantitative and qualitative research paradigms. From these disputes, purists have emerged on both sides. Quantitative purists express assumptions that are consistent with a positivist philosophy, whereas qualitative purists (i.e., post-positivists, post-structuralists,…

Calls for debaters and judges to understand methodological analysis used in debate rounds. Examines several argumentation and debate texts for approaches to the use of statistical evidence. Offers suggestions for handling methodological challenges and a method for weighing such arguments. (PD)

Emphasis is upon the estimation and classical hypotheses testing of economic learning-regression models which may involve systems of equations, collinear regressors, interactions, and distributional effects, when there are errors in the variables and where measures of learning may be ordinal. (Author)

Relates office automation to its application in libraries, discussing computer software packages for microcomputers performing tasks involved in word processing, accounting, statistical analysis, electronic filing cabinets, and electronic mail systems. (EJS)

Urges that libraries become involved in providing access to machine-readable data files, such as statistics gathered by government agencies, research data gathered by scholars, and economic or business data, and describes how the Lawrence University Library developed a data collection and supportive services. Twenty-eight sources are appended.…

Presents several multivariate analyses of variance (MANOVA) test procedures. Discusses guidelines for choosing an overall MANOVA test statistic and post hoc tests that determine the dependent variable or variables responsible for any significant effects. Concludes that guidelines based on recent comparisons of the various test statistics be used.…

After considering how an estimate of learning potential (ELP) is derived from the System of Multicultural Pluralistic Assessment, shows that most of the characteristics of ELP deemed to be important are generated by statistical artifacts. Notes that the fundamental assumption underlying ELP is without supportive evidence. (Author)

The use of cost indexes is encouraged as a method of controlling cost and aiding planning. The use of analysis of variance is encouraged as a method of controlling cost standards. (MSE)

Asserting that the development of educational theory requires scientific study based on comparative data, the author analyzes the objections to data comparison between societies and finds this to be a nonissue. This is the Presidential Address, delivered to the Comparative and International Education Society, Mexico City, March 1978. (SJL)